問題一覧
1
I was so gung ho, so full of energy, so opinionated and so arrogant, to be honest, in those days.
2
as an academic, I didn't really know what I got myself into.
3
But my academic background really hadn't really prepared me that well for what was about to come.
4
the luxury academic assumptions, like machines fail by stopping,or failures are not correlated turned out to be completely unrealistic.
5
I really couldn't pass up the opportunity to build the world's largest distributed system,that was going to be used by almost everyone.
6
I gave you advice to break up your applications into tiers.
7
There is this famous quote that always gets attributed to me.
8
I had really big shoes to fill,but given that actually nobody really knew what the cto job entails, my arrogance got the better of me.
9
I got the opportunity to talk to you and to sort of lay down principles of what I thought was going to be the next generations of development,
10
in this first presentation,I laid out sort of four different categories in which I think I gave you advice.
11
why don't we take a quick look back at it, how well did all of this turn out?
12
I think, you know, continuous integration and continuous deployment and things like that, they're just, everyone's doing that today.
13
Not necessarily because I told you to, but I'm very happy to see that those things made it into everybody's thinking.
14
There a few things that were maybe a little bit more controversial,or took a bit more time.
15
Now, decomposing into small building blocks,loosely coupled, stateless was… at Amazon, this is something we've been doing all the time,and it allowed us to scale and be reliable and evolve pieces independently and things like that.
16
However, it was stepping in the middle of some form of religion, right?
17
It is very hard to build a nano or microservice if you run over rails, and as such you automatically choose for a monolith.
18
architecture is something you do based on the requirements that you have, or maybe sometimes,you do it based on the technology you decided to use.
19
We build it for customers. And whether those customers are internally your business, or they are really your customers externally, you build it for some reason.
20
And as such, you need to collaborate closely with the business and build your architecture in such a way that they can have control over that, one way or another.
21
of course, the business will say everything needs to be 100% reliable.
22
And definitely that point, as I said,now we can architect with cost in mind was blatantly ignored by everyone because suddenly, suddenly you could do all these things.
23
it wasn't until last year that we really saw a desire to start to think and integrate cost into your architectures.
24
Unfortunately, the quote is always incomplete.
25
“The trouble with artificial intelligence is that computers don't give a damn.” But we do. And that's what really grounds us.
26
I want to talk about something that we are prioritizing above all else. That is security.
27
because this is not just a destination, a stand down, a one milestone thing, we’ll never be done.
28
data governance takes on an even more critical,central, important role.
29
we are introducing updates to prevent everything from oversharing, risky use of AI, such as, malicious intent detection, prompt injections, and misuse of protected material.
30
At the end of the day, though, we recognize that when it comes to security,is fundamentally a team sport.
31
it’s rapidly becoming an organizing layer for work and how work gets done.
32
every IT department will have a control system to manage, secure, and measure impact.
33
AI will do for knowledge work. It's all about increasing value and reducing waste.
34
and now they use Copilot to be able to synthesize everything that happened and create that first draft.
35
That means analysis that took weeks, is just taking a day.
36
And now they use Copilot to determine which contracts need to be renewed, which need to be scrapped,and to keep track of all the expiry dates.
37
The more employees use Copilot and Copilot agents, the faster they will be able to realize value.
38
Those systems become more complex over time.
39
None of these services are simple under the covers anymore.
40
You know, survival isn't something that you leave to chance.
41
how we do it safely, securely, and simply, you know, Simplexity.
42
A very relevant law in this context is Tesler’s law, yeah?
43
Complexity can neither be created nor destroyed.
44
He was mostly concerned about user interfaces and the functionality behind it,
45
But it was really fundamental to think about the fact that complexity is a given.
46
Do you get many customers even on cold days like this?
47
Actually ,hot drinks are more popular when its cold,so its a busy time for us.
48
That makes sense
49
I stop by here pretty often on my way to work.
50
Do you work around here?
51
Yes,just a few blocks away.
52
It's a cozy area to work in,
53
Is it usually busy here in the morning.
54
Yes,it gets quite busy,especially on weekdays.
55
I'm glad you like it.
56
I was so excited that I couldn't stop talking.
57
It doesnt matter.
58
Let's call it a day.
59
My aunt always cheers me up.
60
How did you like movie.
61
I have nothing to do with the accident.
62
Do you come here often .
63
The medicine i took hasn't taken effect yet.
64
These shoes are light and comfortable.
65
His news made me happy.
66
There is a 5 hour time difference.
67
I lay on my back and looked at the stars.
68
You can use my charger if you like.
69
Are you john ,by any chance.
70
I'd like to sleep on it.
71
I'm a bit nervous.
72
she was speaking highly of you.
73
No wonder.
74
I really enjoyed talking with you.
75
I used to stay up all night.
76
We have to make the deadline.
77
I couldn't have done it without your help.
78
I'm glad everything worked out.
79
Let me walk with you to the bus stop.
80
About what time are you coming?
81
After you.
82
I wonder if the store is open today.
83
I hurt my shoulder.
84
I need to renew my driver's license next month.
85
Could you do this for me.
86
That's all for today.
87
and indeed, complexity for many of us is inevitable.
88
Those systems become more complex over time, which is a good thing because in general,we're adding more functionality to it.
89
None of these services are simple under the covers anymore.
90
There's some stumblings along the way,
91
One of those things is absolutely eventual consistency in Amazon S3.
92
That was the prevalent thinking—that it was the only way we could make systems reliable and survive failures.
93
it would turn out that the bucket wouldn't be available yet.
94
So, many of our customers tried to implement strong consistency over S3,
95
the right thing to do there was not to remove that complexity,but to move it where it should be.
96
It's moving the complexity to where it needs to be.
97
we could use automated reasoning to prove to ourselves that, actually, everything was correct.
98
counting components doesn't necessarily mean how chaotic your system is,
99
A bicycle has more components, but it is the simplest form from a holistic point of view.
100
Keeping things simple while adding complexity requires discipline.
英語2
英語2
u t · 200問 · 1年前英語2
英語2
200問 • 1年前英語3
英語3
u t · 200問 · 1年前英語3
英語3
200問 • 1年前英語4
英語4
u t · 202問 · 1年前英語4
英語4
202問 • 1年前英語5
英語5
u t · 200問 · 1年前英語5
英語5
200問 • 1年前セキュリティ31-40
セキュリティ31-40
u t · 18問 · 1年前セキュリティ31-40
セキュリティ31-40
18問 • 1年前英語６
英語６
u t · 201問 · 1年前英語６
英語６
201問 • 1年前英語7
英語7
u t · 199問 · 1年前英語7
英語7
199問 • 1年前会社利用
会社利用
u t · 46問 · 1年前会社利用
会社利用
46問 • 1年前問題一覧
1
I was so gung ho, so full of energy, so opinionated and so arrogant, to be honest, in those days.
2
as an academic, I didn't really know what I got myself into.
3
But my academic background really hadn't really prepared me that well for what was about to come.
4
the luxury academic assumptions, like machines fail by stopping,or failures are not correlated turned out to be completely unrealistic.
5
I really couldn't pass up the opportunity to build the world's largest distributed system,that was going to be used by almost everyone.
6
I gave you advice to break up your applications into tiers.
7
There is this famous quote that always gets attributed to me.
8
I had really big shoes to fill,but given that actually nobody really knew what the cto job entails, my arrogance got the better of me.
9
I got the opportunity to talk to you and to sort of lay down principles of what I thought was going to be the next generations of development,
10
in this first presentation,I laid out sort of four different categories in which I think I gave you advice.
11
why don't we take a quick look back at it, how well did all of this turn out?
12
I think, you know, continuous integration and continuous deployment and things like that, they're just, everyone's doing that today.
13
Not necessarily because I told you to, but I'm very happy to see that those things made it into everybody's thinking.
14
There a few things that were maybe a little bit more controversial,or took a bit more time.
15
Now, decomposing into small building blocks,loosely coupled, stateless was… at Amazon, this is something we've been doing all the time,and it allowed us to scale and be reliable and evolve pieces independently and things like that.
16
However, it was stepping in the middle of some form of religion, right?
17
It is very hard to build a nano or microservice if you run over rails, and as such you automatically choose for a monolith.
18
architecture is something you do based on the requirements that you have, or maybe sometimes,you do it based on the technology you decided to use.
19
We build it for customers. And whether those customers are internally your business, or they are really your customers externally, you build it for some reason.
20
And as such, you need to collaborate closely with the business and build your architecture in such a way that they can have control over that, one way or another.
21
of course, the business will say everything needs to be 100% reliable.
22
And definitely that point, as I said,now we can architect with cost in mind was blatantly ignored by everyone because suddenly, suddenly you could do all these things.
23
it wasn't until last year that we really saw a desire to start to think and integrate cost into your architectures.
24
Unfortunately, the quote is always incomplete.
25
“The trouble with artificial intelligence is that computers don't give a damn.” But we do. And that's what really grounds us.
26
I want to talk about something that we are prioritizing above all else. That is security.
27
because this is not just a destination, a stand down, a one milestone thing, we’ll never be done.
28
data governance takes on an even more critical,central, important role.
29
we are introducing updates to prevent everything from oversharing, risky use of AI, such as, malicious intent detection, prompt injections, and misuse of protected material.
30
At the end of the day, though, we recognize that when it comes to security,is fundamentally a team sport.
31
it’s rapidly becoming an organizing layer for work and how work gets done.
32
every IT department will have a control system to manage, secure, and measure impact.
33
AI will do for knowledge work. It's all about increasing value and reducing waste.
34
and now they use Copilot to be able to synthesize everything that happened and create that first draft.
35
That means analysis that took weeks, is just taking a day.
36
And now they use Copilot to determine which contracts need to be renewed, which need to be scrapped,and to keep track of all the expiry dates.
37
The more employees use Copilot and Copilot agents, the faster they will be able to realize value.
38
Those systems become more complex over time.
39
None of these services are simple under the covers anymore.
40
You know, survival isn't something that you leave to chance.
41
how we do it safely, securely, and simply, you know, Simplexity.
42
A very relevant law in this context is Tesler’s law, yeah?
43
Complexity can neither be created nor destroyed.
44
He was mostly concerned about user interfaces and the functionality behind it,
45
But it was really fundamental to think about the fact that complexity is a given.
46
Do you get many customers even on cold days like this?
47
Actually ,hot drinks are more popular when its cold,so its a busy time for us.
48
That makes sense
49
I stop by here pretty often on my way to work.
50
Do you work around here?
51
Yes,just a few blocks away.
52
It's a cozy area to work in,
53
Is it usually busy here in the morning.
54
Yes,it gets quite busy,especially on weekdays.
55
I'm glad you like it.
56
I was so excited that I couldn't stop talking.
57
It doesnt matter.
58
Let's call it a day.
59
My aunt always cheers me up.
60
How did you like movie.
61
I have nothing to do with the accident.
62
Do you come here often .
63
The medicine i took hasn't taken effect yet.
64
These shoes are light and comfortable.
65
His news made me happy.
66
There is a 5 hour time difference.
67
I lay on my back and looked at the stars.
68
You can use my charger if you like.
69
Are you john ,by any chance.
70
I'd like to sleep on it.
71
I'm a bit nervous.
72
she was speaking highly of you.
73
No wonder.
74
I really enjoyed talking with you.
75
I used to stay up all night.
76
We have to make the deadline.
77
I couldn't have done it without your help.
78
I'm glad everything worked out.
79
Let me walk with you to the bus stop.
80
About what time are you coming?
81
After you.
82
I wonder if the store is open today.
83
I hurt my shoulder.
84
I need to renew my driver's license next month.
85
Could you do this for me.
86
That's all for today.
87
and indeed, complexity for many of us is inevitable.
88
Those systems become more complex over time, which is a good thing because in general,we're adding more functionality to it.
89
None of these services are simple under the covers anymore.
90
There's some stumblings along the way,
91
One of those things is absolutely eventual consistency in Amazon S3.
92
That was the prevalent thinking—that it was the only way we could make systems reliable and survive failures.
93
it would turn out that the bucket wouldn't be available yet.
94
So, many of our customers tried to implement strong consistency over S3,
95
the right thing to do there was not to remove that complexity,but to move it where it should be.
96
It's moving the complexity to where it needs to be.
97
we could use automated reasoning to prove to ourselves that, actually, everything was correct.
98
counting components doesn't necessarily mean how chaotic your system is,
99
A bicycle has more components, but it is the simplest form from a holistic point of view.
100
Keeping things simple while adding complexity requires discipline.